The data had subjects.
After cycles, the model
Points should appear close to, and randomly distributed around, the (dashed) identity-line.
*Measures of bias. Others are measures of imprecision.
^Same units as observations and predictions. Others are squared or dimensionless.
Residuals should be randomly distributed around zero, with no apparent trend. For how to interpret these plots, please review our website. Weighted prediction errors are weighted by the observation standard deviation, i.e. the error polynomial.
The model included the following primary parameters
Probability distributions, dashed lines are limits.
The following table lists the final support points, in order of decreasing probability.